Professional Documents
Culture Documents
Energy
journal homepage: www.elsevier.com/locate/energy
a r t i c l e i n f o a b s t r a c t
Article history: The integration of renewable electricity generation poses challenges for today's energy system. Since
Received 4 March 2021 renewable potentials are geographically unevenly distributed, solutions to overcome bottlenecks such as
Received in revised form grid congestion are necessary to raise as much renewable potentials as possible. In specific, a part of
19 May 2021
Austria is characterised with high photovoltaic installation inquiries and low electricity demand. Elec-
Accepted 19 June 2021
tricity grids are designed to cope with today's loads, therefore the integration of inquired photovoltaic
Available online 22 June 2021
would overstrain the built infrastructure. In this paper we investigate how power-to-gas can ease this
photovoltaic related grid strain. The feed-in of hydrogen into natural gas transmission pipelines and the
Keywords:
100 % RES
usage of hydrogen and biogas from local fermentation plants for methanation feeding synthetic natural
Electricity grid congestion relief gas (SNG) into local distribution grid is assessed in various scenarios. Based on the technical assessment
Levelised costs of gas results, an economic evaluation of each scenario is performed to determine generation costs of hydrogen
High degree of renewable penetration or SNG. Depending on the electrolysers location in the electricity grid, the installable photovoltaic power
Multi energy system modelling and can be increased significantly without causing electricity grid congestion. Costs for hydrogen and SNG
simulation vary across each scenario, mainly influenced by CapEx for electrolyser and methanation as well as
electricity purchasing costs.
© 2021 Elsevier Ltd. All rights reserved.
https://doi.org/10.1016/j.energy.2021.121307
0360-5442/© 2021 Elsevier Ltd. All rights reserved.
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
lack of local consumers followed by possible transmission or dis- 1.1.1. Single power-to-gas facility research
tribution grid restrictions if a spatial balance of renewable gener- Using excess wind energy for a single power-to-gas facility,
ation is necessary. To avoid grid congestion, renewable energy Simonis and Newborough [10] investigate a power-to-gas deploy-
sources must be curtailed as ultima-ratio-measure to avoid supply ment scenario in the region of Emden, Germany. The authors pro-
outages [4,8]. Other solution options as an alternative to curtail- vide a detailed technical power-to-gas deployment plan, keeping
ment due to grid restrictions are: grid expansion, sector-coupling or pace with local renewable energies expansion. Hydrogen is either
energy storage. fed into the local natural gas grid or used for methanation by uti-
In order to address the challenges mentioned above, we aim to lising carbon dioxide from a local fermentation plant. The authors’
research also shows that electrolyser utilization increases
Fig. 1. Technical energy potential of RES per Austrian district, derived from Ref. [7] Image size: 2 columns.
2
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
significantly within five years due to an increased availability of electricity grid consists of a 29 busbar model and the natural gas
renewable energy [10]. grid of 79 nodes. Providing an alternative to curtailment, excess
Kopp et al. [16] provide a technical and economic evaluation of wind can be used in power-to-gas units feeding hydrogen or syn-
the already operational 6 MW electrolyser, located in “Energiepark thetic natural gas into gas grids [23].
Mainz”. The technical evaluation investigates the power-to-gas Similar to the research described above, Jentsch and Trost [24]
process efficiency. The economic analysis focuses on various stra- investigate the role of power-to-gas with different degrees of
tegies for electricity procurement, showing that participation in the renewable energy generation in Germany. They use a simplified
german secondary control reserve market can significantly improve spatially and time resolved energy model for Germany, depicting 18
the facility's economic efficiency [16]. defined regions in Germany, to determine an optimized dispatch of
The influence of electricity prices and the usage of produced sector coupling and storage options [24].
hydrogen is assessed by Walker et al. [17]. The authors use various To the best of our knowledge, the effect of power-to-gas on
electricity price limits to activate and deactivate the power-to-gas energy transmission or distribution infrastructure is hardly inves-
facility and analyse corresponding operating costs for various sce- tigated. Since they concentrate their research on power-to-gas
narios. Furthermore, they assess potential revenue sources such as implementation on national state level, Clegg and Mancarella as
replacing hydrogen from steam methane reformation, using well as Jentsch and Trost include a very simplified depiction of
hydrogen for renewable ethanol production, and participating in national energy infrastructure in their research. Their generic
cap and trade carbon dioxide emission trading [17]. infrastructure depiction doesn't allow for any deep analysis of ef-
A framework for locating and modelling the flexibility potentials fects caused by power-to-gas facilities on the built energy infra-
considering temporal and spatial resolution as well as load and structure or on the implementation of RES potentials.
generation in middle voltage distribution systems is introduced by In this paper, we aim to close previously described scientific gap
Henni et al. [15]. The authors demonstrate their approach by by investigating the effects of power-to-gas on a rural distribution
assessing a potential future single power-to-gas facility. The facil- grid section in Austria, with a high degree of potential future
ities locations are selected based on the developed framework. renewable penetration. Furthermore, we aim to bridge between
Local middle voltage electricity and natural gas grids are consid- both before mentioned scientific fields by combining technical
ered. However, the capacity of the grids has to be estimated. macro results with an economic assessment of potentially
Considering avoiding reimbursements or grid expansion, profit- deployable power-to-gas facilities. Following research questions
ability could be achieved in the future [15]. are to be investigated in this paper:
1.1.2. Macro perspective research of power-to-gas - To what extend can photovoltaic be expanded at certain sub-
Estermann et al. [9] investigate the feasibility of power-to-gas to stations, using variable power-to-gas power at different loca-
absorb surplus power from electricity distribution networks grids tions within a certain electricity- and natural gas grid section?
in southern Germany. Furthermore, they include biomass poten- - What are the generation costs for hydrogen and synthetic nat-
tials as a carbon dioxide source for the production of synthetic ural gas?
natural gas. Their focus is on low-voltage distribution grid relief, - What are the effects of power-to-gas sizing and mode of oper-
however, no real-life grids are considered. A total installed power- ation on the economics of the power-to-gas facility?
to-gas capacity of 370 MW could capture 20% of excess solar in
2025 [9]. To answer above described research question we have struc-
A techno-economic assessment of the future role of power-to- tured our paper as followed: The following subchapter describes
gas on both regional and local levels in Baden-Württemberg is upcoming challenges in the examined area. In Section 2 the
assessed by McKenna et al. [18]. A detailed assessment of potential methodology is described. Potential solution scenarios and their
carbon dioxide sources, local renewable generation, as well as corresponding technical and economic results are disclosed in
power-to-gas potentials are presented. The authors expect a cost- Section 3 and 4 respectively. Technical and economic results are
covering operation of power-to-gas in 2030. Any grid restrictions discussed in Section 5, followed by a conclusion and an outlook for
are sparsely considered [18]. potential further research in Section 6.
Guandalini et al. [19] investigated the long term power-to-gas
potential on the national Italian scale if the whole Italian renew- 1.2. Problem description
able potentials are exploited. Five percent of Italy's natural gas
consumption or seven percent of Italy's national fuel consumption The examined part of Austria is sparsely populated without
could be replaced [19]. Similar research is carried out by Bailera large industrial energy consumers but has significant photovoltaic
et al. [20] in Spain, stating a power-to-gas potential between 7 and potentials as can be seen in Fig. 2. Presently, the distribution grid
19,5 GW, depending on excess renewable energy in 2050 [20]. operator is faced with requests for photovoltaic installations that
However, to decarbonise the Spanish electricity sector and indus- are more than double the power of the current 110 kV distribution
trial combined heat and power units, a power-to-gas capacity be- grid capacity (refer to orange lines in Fig. 2). In Fig. 2, the examined
tween 80 and 90 GW is necessary [21]. area including its infrastructure is depicted. Due to anonymization
Greiml et al. [22] demonstrate the positive systemic effects of reasons, each substation is named with a synonym.
power-to-gas taking a federal state's energy infrastructure into It can be seen that substations (SS) with high photovoltaic re-
account. However, the work didn't focus on the power-to-gas fa- quests such as SS South and Rhine are connected to less powerful
cilities location in detail, its economics, and the influence of distribution grid lines compared to other substations. At SS Drava
renewable potentials tap ability [22]. and SS North, the 110 kV distribution grid is connected to the
Clegg and Mancarella [23] identified a lack of research focusing 380 kV transmission grid (refer to red lines in Fig. 2). Adjacent to
on both network implications and benefits of power-to-gas on both the north and west, major natural gas pipelines (NGP) are
electricity and natural gas grids. To overcome before mentioned tangential to the examined area (refer to blue lines in Fig. 2). Near
gap, the authors introduce a new methodology that has been SS Elbe, South and Rhine several fermentation plants are available
applied on a simplified energy grid of Great Britain to investigate as potential carbon dioxide sources. In order to realise as much
the impact of power-to-gas on both electricity and gas grids. The requested photovoltaic power as possible solutions to overcome
3
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
grid congestions are necessary. Following solution scenarios are energy can be transferred across energy carriers via sector coupling
investigated in this paper: options at each node (Fig. 3 left image). Sector coupling options
therefore influence cross sectoral energy carriers.
Installation of power-to-gas electrolyser at SS South with In the current version of HyFlow, the direct current power load
hydrogen feed-in into NGP 2. flow calculation has been upgraded with an alternating power load
Installation of power-to-gas electrolyser at SS North with flow calculation tool (Matpower). Matpower allows for the calcu-
hydrogen feed-in into NGP 1. lation of nodal prices as well as optimal power flow calculations
Installation of power-to-gas electrolyser at SS South. Produced [30].
hydrogen as well as biogas as carbon dioxide source from local The control of hybrid elements depends on the electrical re-
fermentation plants is used for methanation. Produced methane sidual load (RLControl, Power(t) - Eqs. (2) and (1)) of a user-defined
is being feed into local natural gas distribution grid. subsection of the whole depicted area (see Fig. 4 and Eqs. (2-
1)e(2-3)). The described calculation is valid for Pelectrolyser smaller
For each scenario, we investigate to what extend the requested than RLControl, Power. If RLControl, Power exceeds PElectrolyser, the elec-
photovoltaic power can be implemented into the electricity grid at trolyser is operated at its maximum rated power. This approach is
each substation without causing any grid congestion using power- similar to other approaches described in Ref. [9] or [10] to absorb
to-gas at the substations mentioned above. Based on the technical excess renewable generation.
assessment, the levelised costs of gas (LCOG) are to be calculated. X
€
Currently, the guideline G31 issued by Osterreichische Ver- RLControl; Power ðtÞ ¼ PRes;Power;i ðtÞ (2-1)
einigung für das Gas-und Wasserfach allows a maximum hydrogen
content of four percent of the volume in natural gas grids [25]. RLControl; Power ðtÞ 0 ∧ RLControl; Power ðtÞ Pmaxelectr:
However, an increase of the hydrogen content to ten percent of the
volume and further can be expected in the future [26e28]. There- (2-2)
fore, for this paper, the maximum amount of hydrogen in NGP 1 and
2 is defined at ten percent of the volume. PElectrolysis ¼ RLControl; Power ðtÞ (2 e3)
2. Methodology
2.1. Technical assessment
For the load flow calculations within this work, we use the self-
developed software framework HyFlow [22,29]. HyFlow aims for For each scenario, the process depicted in Fig. 5 is manually
load flow based design and operation simulation of flexibility op- performed to determine the maximum installable photovoltaic and
tions such as storage, sector-coupling or demand responsive pro- corresponding electrolyser power. Photovoltaic power iteration
cesses, in multi-energy-systems. In order to allow the consideration starts from confirmed photovoltaic power (current maximum) and
of a broad range of energy system case studies, HyFlow depicts is increased incrementally up to the confirmed, plus requested
three cell levels with various depths of detail for each energy carrier photovoltaic power (refer to “no” path at green diamond in Fig. 5).
considered. For example, level 1 cells can represent a 110 kV dis- In case of grid overloads (refer to “yes” path at green diamond in
tribution grid, level 2 cells a 220 kV transmission grid, and one level Fig. 5), it is evaluated if the electrolyser's power was increased
3 cell a 380 kV transmission grid. Slack nodes allow energy to be before the current simulation. If it has been increased (refer to “yes”
transferred between network levels (Fig. 3 right image). If available, path at blue diamond in Fig. 5), the electrolyser doesn't avoid any
4
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
Fig. 3. Various network levels in combination with cellular approach Image size: 2 columns.
grid overloads, therefore, a maximum PV power has been deter- 2.2. Economic assessment
mined in the previous HyFlow simulation. If the electrolyser's po-
wer wasn't increased (refer to “no” path at blue diamond in Fig. 5) To enable a simplified economic comparison between scenarios,
before the current simulation, the electrolyser's power is increased the levelised costs of gas are calculated based on technical and
and a HyFlow simulation determines if grid overloads still occur. economic parameters for each scenario. To transform one-time
As a result, various electrolyser and photovoltaic powers are capital expenditure (CapEx) into annual payments, the annuity
determined. For each of the investigated scenarios, further exami- method, according to Eqs. (2)e(4) is used [31]. For the calculation of
nations are performed on how produced hydrogen could be used specific CapEx, two annuity factors are calculated. One considers a
and if limitations occur. 12-year period (AF12) for an electrolyser and related planning and
The local distribution grid operator provided us residual loads of construction expenses and the second annuity factor considers a
each substation and photovoltaic generation profile both for one 30-year period (AF30) for pipeline infrastructure, storage,
year in 15 min resolved time-steps as well as technical data to compressor, CO2 separation, and methanation. For both discounting
model power and gas grids. factors an interest rate of four percent is considered. Costs for
5
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
Fig. 5. Determination of maximum photovoltaic and electrolyser power Image size: 1,5 columns.
6
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
Bertuccioli et al. [34] expect PEM electrolysers specific CapEx in CapEx OpEx
a range between 480 and 1270 V/kW, whereas Butler and Spliethoff DN 150 PN 70 300 V/m 1 V/(m a)
[35] expect CapEx within a range between 500 and 1400 V/kW in DN 150 PN 6 150 V/m 1 V/(m a)
2025. Both expect OpEx excluding electricity within a range be-
tween two and five percent of CapEx per year [34,35]. Tremel [36]
includes several studies in PEM CapEx cross-down curves, ranging for low pressure biogas storage (BGS) range between 12 and 53
from 400 to 900 V/kW in 2025. Thema et al. [37] expect PEM € hler [43].
V/m3, according to Do
electrolyser OpEx to be at about 900 V/kW in 2025. The electrolyser
efficiency of current projects is up to 77% [37]. Schiebahn et al. [38]
2.8. Compressor
expect a potential future PEM electrolyser efficiency up to 74%.
Manufacturers of PEM electrolysers such as H-TEC and Siemens
Costs for hydrogen compressors vary within a wide range, ac-
Energy specify the plant efficiency of their products between 74
cording to van Leuween et al. [44]. CapEx range between 144 and
and 75,5% [39,40].
18,500 V/kW, indicating that costs for compressors depend highly
on their application. Most values spread around 1500 V/kW for
2.4. Methanation compression up to 70 bar. OpEx range between 1,5 and 4% of CapEx
[44]. Costs for methane and biogas compressors were defined as
Gorre et al. [41] expects CapEx of 295e375 V/kWSNG and fixed shown in Table 3 according to Refs. [26,44].
OpEx of five percent of CapEx as well as a variable OpEx of 0.63
V/(MWel h) in 2030 [41]. A study from Frontier Economics expect 2.9. Planning and construction
CapEx of 432e756 V/kWCH4 in 2030 and annual OpEx of four
percent of CapEx. The methanations efficiency is, based on several For planning and construction of electrolyser and methanation
other studies defined in a range between 78 and 83% [42]. To Gorre et al. [41] consider planning costs of 100e160 tV. For con-
convert V/kWSNG and V/kWCH4 into V/kWElectrolyser CapEx they struction, 10% of CapEx are considered [41].
have to be multiplied with electrolyser and methanation efficiency.
Based on selected efficiencies of an electrolyser (75%) and metha-
2.10. Electricity purchase
nation (79%), CapEx ranges between 175 and 445 V/kWel [42].
A study from Kost et al. [45] regarding electricity generation
2.5. CO2 separation costs provides cross-down curves for various renewable energies
until the year 2035. For the year 2025, electricity generation costs
€ hler [43] provides CapEx as well as fixed and variable OpEx
Do for open space photovoltaic in Germany are forecasted to range
for various CO2 separation technologies such as pressure water between 3 and 5,5 Vcent/kWh [45]. Considering higher solar ra-
scrubbing, amine gas treating, and pressure swing adsorption. diation in the south of Austria compared to Germany, electricity
Based on costs for the mentioned technologies, each cost category generation costs can be assumed to be at the lower end or below
is averaged. For several processing capacities, a cost function is the described range [46]. The International Renewable Energy
determined via regression (result see Table 3) [43]. Agency [47] expects the global levelised costs of electricity for
photovoltaics to drop to a range of 0,02 to 0,08 US$/kWh until the
2.6. Natural gas and hydrogen pipeline year 2030. For this study, costs for electricity purchases of 30
V/MWh are considered. Network fees for electricity grid are
Average costs for both natural gas and hydrogen transmission considered amounting to 1,15 VCent/kWh excluding value-added
and distribution pipelines are provided by van Leuween et al. [44]. tax [48,49]. Electricity purchasing prices are defined considering
The authors provide different cost estimates based on various near term future PV generation costs. If electricity is purchased
studies depending on the pipeline surroundings, as displayed in from open market, the price might fluctuate depending on
Table 1. Hydrogen pipelines require advanced technology compared addressed market and season.
to natural gas pipelines, resulting in higher CapEx [44].
OpEx are defined as 2% of CapEx [44]. In Ref. [26] costs for 2.11. Selected costs for economic assessment
natural gas pipelines were defined as shown in Table 2.
In Table 3, the costs for necessary facilities based on literature
2.7. Hydrogen and biogas storage overview for the economic assessment of each scenario are
displayed.
Costs for storage of hydrogen and biogas are widely spread
depending on pressure, size, and technology. Gorre et al. [41] 3. Scenarios
expect hydrogen storage (HS) costs of 75 V/Nm3 in the year 2030
[41]. Van Leuween et al. [44] define the following costs for steel In this section the three investigated scenarios are explained.
tanks based on several literature sources: CapEx range between 23 The local distribution grid operator provided us residual loads of
and 195 V/m3, and OpEx between 0,5 and 2,5% of CapEx [44]. CapEx each substation and photovoltaic generation profile both for one
year in 15 min resolved time-steps as well as technical data to
model power and gas grids.
Table 1
Pipeline costs according to van Leuween [44].
3.1. Scenario 1: hydrogen feed into NGP 2
Rural Medium Urban
CapEx methane pipeline [V/m] 100 300 500 As displayed in Fig. 7, an electrolyser is placed at SS South.
CapEx hydrogen pipeline [V/m] 350 450 550
Produced hydrogen is fed into NGP 2. Two modes of operation for
7
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
Table 3
Defined costs for economic assessment.
the electrolyser are investigated to determine a suitable mode of approximately 3400 Nm3/h. A central electrolyser is placed at
operation to increase electrolyser utilization: substation South, operated like scenario 1 (see Fig. 9) (see Fig. 8).
This scenario is divided into sub-scenarios, as displayed in
Electrolyser is activated once residual load of substations South Fig. 10.
and Rhine turns negative (refer to Fig. 4). To overcome different production profiles between volatile
Electrolyser is activated once the residual load of all substations photovoltaic powered electrolyser and steady fermentation plants,
except Drava, North and West turns negative. storage options are investigated for both hydrogen and biogas in
technical terms such as storage capacity as well as economic terms.
In Fig. 11 different production profiles for both hydrogen and biogas
3.2. Scenario 2: hydrogen feed into NGP 1 are displayed, displaying a need for storage solutions.
To avoid biogas flaring or curtailment of fermentation plants in
In scenario 2, an electrolyser is placed at substation North and case no hydrogen is available from storage or photovoltaic gener-
produced hydrogen is fed into NGP 1. The electrolyser is operated ation, two further cases for each sub-scenario are considered. These
like in scenario 1 with the addition of SS Maas in operating strategy include CO2 separation (CO2 sep) from biogas to produce methane,
1. to be fed into the local distribution grid, or electricity purchase (EP)
to operate the electrolyser with purchased electricity and provide
3.3. Scenario 3: Methane/SNG feed into local natural gas hydrogen for methanation.
distribution grid
4. Results
This scenario takes into consideration local fermentation plants
in proximity to substations Elbe, South, and Rhine. The total In this section we present technical as well as economical results
amount of raw gas (60% CH4, 40% CO2) available for methanation is for each scenario.
8
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
Fig. 9. Scenario 3 - Methane feed into local distribution grid Image size: 2 columns.
4.1. Scenario 1: hydrogen feed into NGP 2 load of substations South and Rhine (operation strategy #1) is far
more advantageous compared to operating strategy #2 when
4.1.1. Assessment of technical results considering achieved full-load hours. Substations South and Rhine
The electrolysers mode of operation significantly impacts the have a low consumption in combination with high photovoltaic
achievable full-load hours, as displayed in Fig. 12, where achieved potentials. An increased utilization occurs because an over-
full-load hours based on installed photovoltaic and electrolyser production of electricity is achieved earlier, leading to longer
power are shown. Operating the electrolyser based on the residual electrolyser production time. Operating strategy #2 increases
9
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
Fig. 11. Display of different production profiles and storage usage Image size: 1,5 columns.
Fig. 12. Electrolyser full-load hours in dependence on operation strategy Image size: 2 columns.
demand in combination with a marginal increase of photovoltaic “PV inquired þ committed” power contains the sum of photovoltaic
potentials, resulting in a decreasing electrolyser utilization. power installation inquiries. Up to 500 MW of photovoltaic power
Therefore, results based on operating strategy #1 will be displayed could be added to the grid by installing a 227,5 MW electrolyser. A
further onwards. further increase of electrolyser power cannot integrate any further
Fig. 13 displays the maximum photovoltaic power per substat- photovoltaic into the electricity grid due to overloads (refer to
ion that can be added to the electricity grid for certain electrolyser “electricity grid limitation” in Fig. 13), occurring at grid sections
powers. “PV committed” represents the current maximum instal- between SS Maas - North and South - Rhine.
lable photovoltaic power according to distribution grid operator. However, the minimum flowrate in NGP 2 during summer is
10
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
Fig. 13. Maximum installable PV per SS and electrolyser power e scenario 1 Image size: 2 columns.
hydrogen feed-in limit. A hydrogen storage could store excess 35 MW electrolyser 52,5 MW electrolyser
hydrogen during daytime, and feed into NGP 2 during night time. Electrolyser 28 Mio. V 42 Mio. V
Implementing a storage could further increase the installable H2 storage e 2,6 Mio. V
electrolyser power up to approximately 100 MW (refer to “feed-in H2 pipeline 5,5 Mio. V 5,5 Mio. V
limit NGP 2” in Fig. 13). Table 4 contains technical key figures of H2 compressor 645,000 V 967,500 V
Planning and construction 2,96 Mio. V 4,36 Mio. V
scenario 1. It can be seen that an increase in electrolyser power
Produced H2 73,0 GWh 102,3 GWh
results in a decrease of both full-load hours and the ratio between Spec. CapEx 0,050 V/kWhH2 0,053 V/kWhH2
added photovoltaic power compared to electrolyser power (Padded
PV: PEL installed).
Table 6
4.1.2. Assessment of economic results Annual OpEx and electricity costs for scenario 1.
The following tables display calculated CapEx (Table 5) as well as
35 MW electrolyser 52,5 MW electrolyser
OpEx and Electricity costs (Table 6) for two different electrolyser
dimensions. CapEx for electrolyser representing about 75% of total Electrolyser 840,000 V 1,26 Mio. V
H2 storage e 51,600 V
CapEx is the main CapEx driver. OpEx and electricity costs are H2 pipeline 11,000 V 11,000 V
predominantly caused by the electrolyser, therefore about 90% of H2 compressor 24,495 V 36,744 V
specific hydrogen costs are caused by the electrolyser. Spec. OpEx 0,012 V/kWhH2 0,013 V/kWhH2
Specific hydrogen costs of 11,7 and 12,2 VCent/kWhH2 are equal Annual electricity costs 4,04 Mio. V 5,66 Mio. V
Spec. electricity costs 0,055 V/kWhH2 0,055 V/kWhH2
to 4,4 and 4,8 V/kgH2.
4.2.1. Assessment of technical results Due to the power grid limitation explained before, we provide
Fig. 14 displays the maximum photovoltaic power per substat- technical results in Table 7, with comparable electrolyser powers as
ion that can be added to the electricity grid for a certain electrolyser in scenario 1.
power. It can be seen that any increase in electrolyser power Compared to scenario 1, the added photovoltaic power doesn't
doesn't correspond with an increase of installed photovoltaic. Since increase with rising electrolyser power. Therefore the ratio
the electrolyser is located at a substation without any photovoltaic Padded PV: PEL installed is lower compared to scenario 1.
potentials, the electricity has to be transferred via the electricity
grid to the electrolyser. This causes overloads of the electricity grid
section Maas e North. 4.2.2. Assessment of economic results
The following tables display calculated CapEx (Table 8) as well as
annual OpEx and Electricity costs (Table 9) for two different elec-
trolyser dimensions.
Table 4
Specific hydrogen costs of 13,4 and 14,0 VCent/kWhH2 are equal
Technical results of scenario 1.
to 5,3 and 5,5 V/kgH2. Although the absolute CapEx values for the
35 MW electrolyser 52,5 MW electrolyser equipment are lower compared to scenario 1, because of fewer
added PV Power 152 MW 172 MW infrastructure requirements (H2 pipeline and storage), the specific
full-load hours electrolyser 2770 h 2598 h CapEx values are higher due to fewer electrolyser full-load hours.
electricity consumption 97,0 GWh 136,4 GWh Lower full-load hours compared to scenario 1 are caused by
Padded PV: PEL installed 4,3 3,3
including SS Maas into the control area of the electrolyser.
11
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
Fig. 14. Maximum installable PV per SS and electrolyser power e scenario 2 Image size: 2 columns.
Fig. 15. Results electricity purchase for both sub-scenarios Image size: 2 columns.
Fig. 17. Comparison of electrolyser full-load hours between CO2 separation and electricity purchase Image size: 2 columns.
Table 10
Technical results of scenario 2.
13
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
Fig. 19. Technical limits comparison of each scenario Image size: 1,5 columns.
power is greater. Additionally, a smaller electrolyser can achieve electricity demand, the local distribution grid's capacity is not
higher full-load hours, benefitting the electrolysers economic effi- designed to integrate high PV potentials. We believe the described
ciency. The correlation between higher achievable full-load hours approach of using an electrolyser as a grid relief measure can be
and smaller electrolyser power is also observed in Ref. [10], considered in any region with a similar characteristic, provided the
although electricity is sourced from wind power. natural gas grid is accessible and within close proximity. A region
In case grid expansion is considered as an alternative to power- with similar characteristics may include:
to-gas, the grid section from SS Elbe via Rhine and Maas to North,
totaling approximately 100 km length, must be strengthened. High potential of renewable energy sources
CapEx of 600 V/m, based on an Austrian 110 kV grid construction Low transmission capacity of electricity grids
cost analysis can be used to determine grid expansion costs [54]. Low electricity demand
Without a detailed analysis, the CapEx for grid strengthening of 60 Rural, sparsely populated area
Mio.V is comparable to the presented power-to-gas scenarios.
In case a carbon-dioxide source (e.g. from: fermentation plants,
industry) is available, produced hydrogen can be used for metha-
6. Conclusion and future outlook nation and feed-in natural gas grids without restrictions.
The examined scenarios show that an electricity grid relief can
Within this work, we discuss the positive impact of power-to- be achieved using power-to-gas as long as the electrolyser is
gas sector-coupling in terms of increasing renewable photovoltaic correctly positioned. The location as well as the mode of operation
electricity generation in an Austrian region. Current literature influences both the technical and economic efficiency of the elec-
regarding power-to-gas research is mainly focused on either a trolyser. Technically, using a 227,5 MW electrolyser, up to 500 MW
regional or national level power-to-gas potential, or the operation of photovoltaic could be implemented into the current electricity
of a single power-to-gas facility. A scientific gap was identified; grid. Depending on the electrolyser's rated power, between 1650
both mentioned fields of research don't consider the restrictions in full-load hours for a 227,5 MW electrolyser up to 3000 full-load
real-life energy grids in their assessment. To bridge between both hours for a 17,5 MW electrolyser from excess photovoltaic gener-
areas of research we conducted a technical assessment of the in- ation can be achieved. In case electricity purchases are allowed, the
fluence of power-to-gas on an existing energy grid and used tech- electrolyser's full-load hours can be increased up to 4700 h. Smaller
nical results to determine the production costs for renewable electrolyser powers are favourable in terms of both economic and
hydrogen or SNG. technical efficiency. Further research should be carried out if
The examined region is rural, lagging major electricity con- several smaller decentralised electrolysers are favourable
sumers but offers high renewable potentials. Due to the low
15
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
compared to one centralised electrolyser in both technical and excess solar power in electricity distribution networks. 32. In: International
journal of hydrogen energy, vol. 41; 2016. p. 13950e9. https://doi.org/
economic efficiency. Furthermore, the usage of waste heat could be
10.1016/j.ijhydene.2016.05.278.
investigated to increase the electrolysers economic efficiency. [10] Simonis B, Newborough M. Sizing and operating power-to-gas systems to
Based on technical results, such as electrolyser's full-load hours absorb excess renewable electricity. In: International journal of hydrogen
and literature values for possible further cost developments, an energy, vol. 42; 2017. p. 21635e47. https://doi.org/10.1016/
j.ijhydene.2017.07.121.
economic assessment for each scenario was carried out. The eco- [11] Walker Sean B, Mukherjee Ushnik, Fowler Michael, Elkamel Ali. Bench-
nomic assessment shows that generation costs for hydrogen (11,7 marking and selection of Power-to-Gas utilizing electrolytic hydrogen as an
to 14 VCent/kWh) or SNG (14e16,4 VCent/kWh) in Austria are in energy storage alternative. In: International journal of hydrogen energy, vol.
41; 2016. p. 7717e31. https://doi.org/10.1016/j.ijhydene.2015.09.008.
line with literature expectations. The main cost drivers for both [12] van Leeuwen Charlotte, Mulder. Machiel: power-to-gas in electricity markets
renewable hydrogen and SNG are CapEx for electrolyser and dominated by renewables. In: Applied energy, vol. 232; 2018. p. 258e72.
methanation as well as energy purchasing costs. Although costs are https://doi.org/10.1016/j.apenergy.2018.09.217.
[13] Hidrogeno Abengoa, Liquide Air, AkzoNobel, Alstom ECN, elementenergy,
calculated to the best of our knowledge based on various literature eurogas, GERG, HEDNO. Hellenic petroleum, HIA, hydrogenics, inabensa, ITM
sources, deviations from the assumed costs may lead to signifi- power, linde group, moixa technology, NEL hydrogen, nothern power grid,
cantly different generation costs for hydrogen or SNG. NOW, proton on site, PAE RAE OREDT, res AMERICAS, shell, SIEMENS, statk-
raft, statoil, sunfire, TATA chemicals, VANADIS POWER, VATTENFALL, VOITH.
Raw materials for fermentation plants and large-scale photo- In: Commercialisation of energy storage in Europe : a fact-based analysis of
voltaic consume large land areas for plantation and installation, the implications of projected development of the European electric power
potentially leading to land use conflicts. One further step should be system towards 2030 and beyond for the role and commercial viability of
energy storage; 2015. Brussels, Match, https://www.fch.europa.eu/sites/
a life-cycle assessment in order to examine the environmental
default/files/CommercializationofEnergyStorageFinal_3.pdf.
impact of the proposed solution. [14] Bellocchi Sara, Falco Marcello de, Gambini Marco, Manno Michele,
Stilo Tommaso, Vellini Michela. Opportunities for power-to-Gas and Power-
to-liquid in CO2-reduced energy scenarios: the Italian case. In: Energy, vol.
Credit author statement 175; 2019. p. 847e61. https://doi.org/10.1016/j.energy.2019.03.116.
[15] Henni Sarah, Staudt Philipp, Kandiah Balendra, Weinhardt Christof. Infra-
Matthias Greiml: Conceptualization, Methodology, Validation, structural coupling of the electricity and gas distribution grid to reduce
renewable energy curtailment. In: Applied energy, vol. 288; 2021. p. 116597.
Formal analysis, Writing e original draft. Florian Fritz: Methodol- https://doi.org/10.1016/j.apenergy.2021.116597.
ogy, Software, Formal analysis, Visualization. Thomas Kienberger: [16] Kopp M, Coleman D, Stiller C, Scheffer K, Aichinger J, Scheppat B,
Conceptualization, Supervision, Writing e review & editing. Mainz Energiepark. Technical and economic analysis of the worldwide largest
Power-to-Gas plant with PEM electrolysis. 19. In: International journal of
hydrogen energy, vol. 42; 2017. p. 13311e20. https://doi.org/10.1016/
Declaration of competing interest j.ijhydene.2016.12.145.
[17] Walker Sean B, van Lanen Daniel, Fowler Michael, Mukherjee Ushnik. Eco-
nomic analysis with respect to Power-to-Gas energy storage with consider-
The authors declare that they have no known competing ation of various market mechanisms. In: International journal of hydrogen
financial interests or personal relationships that could have energy, vol. 41; 2016. p. 7754e65. https://doi.org/10.1016/
appeared to influence the work reported in this paper. j.ijhydene.2015.12.214.
[18] McKenna RC, Bchini Q, Weinand JM, Michaelis J, Ko €nig S, Ko € ppel W,
Fichtner W. The future role of Power-to-Gas in the energy transition: regional
Acknowledgement and local techno-economic analyses in Baden-Württemberg. In: Applied en-
ergy, vol. 212; 2018. p. 386e400. https://doi.org/10.1016/
j.apenergy.2017.12.017.
The authors thank E-Netze Steiermark for their kind contribu- [19] Guandalini G, Robinius M, Grube T, Campanari S, Stolten D. Long-term power-
tions to this project. to-gas potential from wind and solar power: a country analysis for Italy. 19.
In: International journal of hydrogen energy, vol. 42; 2017. p. 13389e406.
https://doi.org/10.1016/j.ijhydene.2017.03.081.
References [20] Bailera Manuel, Lisbona. Pilar: Energy storage in Spain: Forecasting electricity
excess and assessment of power-to-gas potential up to 2050. Energy 2018;143:
[1] Collins Se an, Deane John Paul, Poncelet Kris, Panos Evangelos, 900e10. https://doi.org/10.1016/j.energy.2017.11.069.
Pietzcker Robert C, Delarue Erik, Gallacho P
ir O, adraig Brian. Integrating short [21] Lisbona Pilar, Frate Guido Francesco, Bailera Manuel, Desideri Umberto. Po-
term variations of the power system into integrated energy system models: a wer-to-Gas: analysis of potential decarbonization of Spanish electrical system
methodological review. In: Renewable and sustainable energy reviews, vol. in long-term prospective. Energy 2018;159:656e68. https://doi.org/10.1016/
76; 2017. p. 839e56. https://doi.org/10.1016/j.rser.2017.03.090. j.energy.2018.06.115.
[2] Hansen Kenneth, Breyer Christian, Lund Henrik. Status and perspectives on [22] Greiml Matthias, Traupmann Anna, Sejkora Christoph, Kriechbaum Lukas,
100% renewable energy systems. Energy 2019;175:471e80. https://doi.org/ Bo € ckl Benjamin, Pichler Patrick, Kienberger Thomas. Modelling, designing and
10.1016/j.energy.2019.03.092. operation of grid-based multi-energy systems. 7-24 Pages. Intl J Sustain Energy
[3] Mancarella Pierluigi. MES (multi-energy systems): an overview of concepts Plan manag 2020;29. https://doi.org/10.5278/ijsepm.3598. International
and evaluation models. In: Energy, vol. 65; 2014. p. 1e17. https://doi.org/ Journal of Sustainable Energy Planning and Management, Vol. 29 (2020)
10.1016/j.energy.2013.10.041. (2020).
[4] Kroposki Benjamin, Johnson Brian, Zhang Yingchen, Gevorgian Vahan, [23] Clegg Stephen, Mancarella Pierluigi. Integrated modeling and assessment of
Denholm Paul, Hodge Bri-Mathias, Hannegan Bryan. Achieving a 100% the operational impact of power-to-gas (P2G) on electrical and gas trans-
renewable grid: operating electric power systems with extremely high levels mission networks. In: IEEE transactions on sustainable energy, 6; 2015.
of variable renewable energy. 2. In: IEEE power and energy magazine, vol. 15; p. 1234e44. https://doi.org/10.1109/TSTE.2015.2424885.
2017. p. 61e73. https://doi.org/10.1109/MPE.2016.2637122. [24] Jentsch Mareike, Trost Tobias, Ho € fling Holger (Mitarb), Jachmann Henning
[5] #mission2030: #mission2030 : Austrian Climate and Energy Strategy. 2018. (Mitarb). Analyse von Power-to-Gas-Energiespeichern im regenerativen
Vienna, September, https://mission2030.info/wp-content/uploads/2018/10/ Energiesystem. Kassel; October 2014. https://www.researchgate.net/
Klima-Energiestrategie_en.pdf. publication/301688145_Analyse_von_Power-to-Gas-Energiespeichern_im_
[6] Gruber Karl Heinz. Umsetzung der #mission2030-Ziele aus Sicht von Oes- regenerativen_Energiesystem.
terreichs Energie. https://iewt2019.eeg.tuwien.ac.at/download/contribution/ [25] €
Osterreichische Vereinigung für das Gas- und WAsserfach: Regeln der OVGW€ -
presentation/327/327_presentation_20190213_173020.pdf. Review date: €
richtlinie G 31 : Erdgas in Osterreich: Gasbeschaffenheit. 2001. Wien.
2021-01-08. [26] €
OVGW, Leoben Montanuniversita €t, Wien WU. DBI Gas- und Umwelttechnik
[7] Sejkora Christoph, Kühberger Lisa, Radner Fabian, Trattner Alexander, GmbH, TU Wien, JKU Linz, ERIG: Greening the Gas : forschungsbericht 2019.
Kienberger Thomas. Exergy as criteria for efficient energy systemsda spatially 2vol. 29; 2020. Vienna, https://www.ovgw.at/media/medialibrary/2020/03/
resolved comparison of the current exergy consumption, the current useful OVGW_JB_forschung19_hi_corr2.pdf.
exergy demand and renewable exergy potential. 4. In: Energies, vol. 13; 2020. [27] Gas Connect Austria. Mit Hochdruck für mehr Wasserstoff e status quo und
p. 843. https://doi.org/10.3390/en13040843. Chancen. https://www.gasconnect.at/aktuelles/news-presse/news/detail/
[8] Doering Michael. Assessment of storage options for reduction of yield losses in News/mit-hochdruck-fuer-mehr-wasserstoff-status-quo-und-chancen.
a region with 100% renewable electricity. In: Energy procedia, vol. 73; 2015. Aktualisierungsdatum: 2021-01-11.
p. 218e30. https://doi.org/10.1016/j.egypro.2015.07.675. [28] Austrian Gas Grid Management AG, Langfristige Planung. Für die Gas Vertei-
[9] Estermann T, Newborough M, Sterner M. Power-to-gas systems for absorbing €
lernetzinfrastruktur in Osterreichfür den Zeitraum 2020e2029. https://www.
16
M. Greiml, F. Fritz and T. Kienberger Energy 235 (2021) 121307
e-control.at/documents/1785851/1811582/LFP19_Bericht_A2_xxxBGG_inkl- doi.org/10.1016/j.apenergy.2019.113594.
Anhang.cleaned.pdf/dd1c0f44-05d4-0cc6-b951-9460e5bfaa8b? [42] Verkehrswende Agora, Energiewende Agora. Frontier economics; perner, jens
t¼1574159767808; 2019. 2019. (mitarb.); unteutsch, michaela (mitarb.); lo € venich, andrea (mitarb.); deutsch,
[29] Bo€ckl Benjamin, Greiml Matthias, Leitner Lukas, Pichler Patrick, matthias (mitarb.); maier, urs (mitarb.). Berlin: Die zukünftigen Kosten
Kriechbaum Lukas, Kienberger Thomas. HyFlowda hybrid load flow- strombasierter synthetischer Brennstoffe; 2018. https://www.agora-
modelling framework to evaluate the effects of energy storage and sector energiewende.de/fileadmin2/Projekte/2017/SynKost_2050/Agora_SynCost-
coupling on the electrical load flows. Energies 2019;12(5):956. https:// Studie_WEB.pdf.
doi.org/10.3390/en12050956. [43] Do€hler, Helmut (Hrsg.). Faustzahlen biogas. 3. Ausg. Darmstadt: KTBL; 2013.
[30] Zimmerman Ray Daniel, Murillo-Sanchez, Edmundo Carlos, Thomas Robert 978-3-941583-85-6.
John. MATPOWER: steady-state operations, planning, and analysis tools for [44] van Leeuwen, Zauner Charlotte. Andreas; Gorre, Jachim (Mitarb.): Innovative
power systems research and education. 1. In: IEEE transactions on power large-scale energy storage technologies and Power-to-Gas concepts after opti-
systems, vol. 26; 2011. p. 12e9. https://doi.org/10.1109/ misation : Report on the costs involved with PtG technologies and their potentials
TPWRS.2010.2051168. across the EU. 30.04. 2018. https://ec.europa.eu/research/participants/
[31] Konstantin Panos. Praxisbuch Energiewirtschaft : energieumwandlung, -trans- documents/downloadPublic?
port und -beschaffung, Übertragungsnetzausbau und Kernenergieausstieg. 4., documentIds¼080166e5ba3ba6a8&appId¼PPGMS.
aktualisierte Auflage. Berlin: Springer Vieweg; 2017 (VDI-Buch). 978-3-662- [45] Kost Christoph, Shammugam Shivenes, Jülch Verena, Nguyen Huyen-Tran,
49823-1. Schlegl Thomas, Henning Hans-Martin (Mitarb). Bett, andreas (mitarb.) :
[32] Raikar Santosh, Adamson Seabron. Renewable energy finance in the inter- stromgestehungskosten erneuerbare energien. Freiburg. https://www.ise.
national context. In: Raikar Santosh, Adamson Seabron, Hrsg, editors. fraunhofer.de/content/dam/ise/de/documents/publications/studies/DE2018_
Renewable energy finance : Theory and practice. London: Academic Press; ISE_Studie_Stromgestehungskosten_Erneuerbare_Energien.pdf; March 2018.
2020. p. 185e220. [46] European Commission. Photovoltaic geographical information system
[33] Wulf Christina, Linßen Jochen. Zapp, petra: Review of power-to-gas Projects in (PVGIS). https://ec.europa.eu/jrc/en/pvgis. Review date: 2021-01-08.
Europe. Energy Procedia 2018;155:367e78. https://doi.org/10.1016/ [47] International renewable energy agency. Future of solar photovoltaic :
j.egypro.2018.11.041. deployment, investment, technology, grid integration and socio-economic
[34] Bertuccioli Luca, Chan Alvin, Hart David, Lehner Franz, Madden Ben, aspects. Abu Dhabi: A Global Energy Transformation Paper; 2019. 978-992-
Standen Eleanor. Study on development of water electrolysis in the EU. 7.2. 9260-156-0.
https://www.fch.europa.eu/sites/default/files/study%20electrolyser_0-Logos_ €
[48] Bundeskanzleramt. Anderung der Systemnutzungsentgelte-verordnung 2018
0_0.pdf; 2014. (SNE-V 2018 - novelle 2020) (idF v. BGBl. II nr. 424/2019 - teil 2). In: BGBl. II
[35] Buttler Alexander, Spliethoff Hartmut. Current status of water electrolysis for Nr. 424/2019; 2019.
energy storage, grid balancing and sector coupling via power-to-gas and [49] Wko OO. € Systemnutzungsentgelte für Strom. https://www.wko.at/service/
power-to-liquids: a review. In: Renewable and sustainable energy reviews, ooe/umwelt-energie/Die-Strom-Netzentgelte-2020-im-Detail.pdf. Review
vol. 82; 2018. p. 2440e54. https://doi.org/10.1016/j.rser.2017.09.003. date: 2021-03-01.
[36] Tremel Alexander. Electricity-based fuels. Cham: Springer International Pub- [50] Austrian gas grid management: marktgebietsdaten. https://platform.aggm.at/
lishing; 2018. 978-3-319-72458-4. vis/visualisation/map.
[37] Thema M, Bauer F, Sterner M. Power-to-Gas: electrolysis and methanation [51] Zauner Andreas, Bo € hm Hans, Rosenfeld Daniel C, Tichler Robert. Innovative
status review. In: Renewable and sustainable energy reviews, vol. 112; 2019. large-scale energy storage technologies and Power-to-Gas concepts after
p. 775e87. https://doi.org/10.1016/j.rser.2019.06.030. optimization. Analysis on future technology options and on techno-economic
[38] Schiebahn Sebastian, Grube Thomas, Robinius Martin, Tietze Vanessa, optimization; 2019. 28.02, https://ec.europa.eu/research/participants/
Kumar Bhunesh, Stolten Detlef. Power to gas: technological overview, systems documents/downloadPublic?
analysis and economic assessment for a case study in Germany. 12. In: In- documentIds¼080166e5c58ae3ff&appId¼PPGMS.
ternational journal of hydrogen energy, vol. 40; 2015. p. 4285e94. https:// [52] Bukold Steffen. Blauer wasserstoff : Perspektiven und Grenzen eines neuen
doi.org/10.1016/j.ijhydene.2015.01.123. technologiepfades. January, https://www.greenpeace-energy.de/fileadmin/
[39] SIEMENS energy. Silyzer 300 : The next paradigm of PEM electrolysis. https:// docs/publikationen/Studien/blauer-wasserstoff-studie-2020.pdf; 2020.
assets.siemens-energy.com/siemens/assets/api/uuid:17cdfbd8-13bd-4327- [53] Kayfeci Muhammet, Keçebaş Ali, Bayat Mutlucan. Hydrogen production. In:
9b26-383e4754bee3/datasheet-final1.pdf. Review date: 2021-01-08. Solar hydrogen production. Elsevier; 2019. p. 45e83. https://doi.org/10.1016/
[40] H-TEC SYSTEMS. PEM-Elektrolyseur e das Bindeglied für Sektorenkopplung B978-0-12-814853-2.00003-5.
und dezentrale H₂-Produktion. https://www.h-tec.com/fileadmin/Content/ [54] Lehner Erich, Schrenk Katharina. Gutachten im Auftrag der OO € Landesholding
PDFs/19022019/H-TEC_SYSTEMS_Datenblatt_Elektrolyseur_ME450_1400_DE. GmbH zum Investitions-und Kostenvergleich FreileitungeErdkabel auf der
pdf. Review date: 2021-01-08. 110 kV-Spannungsbene im la €ndlichen Raum. December. Linz; 2018. https://
[41] Gorre Jachin, Ortloff Felix, van Leeuwen Charlotte. Production costs for syn- www.land-oberoesterreich.gv.at/Mediendateien/Formulare/Dokumente%
thetic methane in 2030 and 2050 of an optimized Power-to-Gas plant with 20UWD%20Abt_US/4_RK_18_Gutachten_OoeLaHol_FINAL_20181221.pdf.
intermediate hydrogen storage. Appl Energy 2019;253:113594. https://
17